In recent years, federal and state environmental agencies have looked to natural resource damages (NRD) claims to restore waterways and land areas harmed by pollution. But NRD concepts are relatively new and not widely understood. Many companies with potential exposure are only beginning to address these exposures in their risk management planning.

This article presents a short introduction to NRD. It first explains what natural resource damages are and are not, then describes how NRD liability is assessed and finally explains why companies facing possible NRD claims should become engaged in the natural resource damages assessment (NRDA) process as early as possible.

What Natural Resource Damages Are and Are Not

The two primary federal statutes supporting assertion of NRD claims are in the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA)1 and the Oil Pollution Act (OPA)2. Many states also have enacted analogous statutes.3

Under these statutes, NRD claims are brought on behalf of the public by federal, state or tribal agencies (trustees) against parties alleged to be responsible for unpermitted releases of oil or hazardous substances that, in turn, have allegedly injured natural resources and resulted in the loss of ecological and human use services those resources provide. “Ecological services” refers to beneficial ecological functions, such as the function of wetlands in protecting against erosion or subsidence. “Human use services” include non-economic benefits people receive from natural resources, such as the recreational benefit of fishing or viewing wildlife.

Natural resource damages are based on the costs of restoring injured natural resources to “baseline”—that is, the condition those resources would have been in had the unpermitted releases not occurred. In addition to these “primary” restoration costs, responsible parties (RPs) also can be assessed “compensatory” restoration costs. These are the costs of providing natural resource services equivalent to those lost before restoration to baseline is complete. Finally, natural resource damages can also include the reasonable costs of assessment.

NRD is limited to publicly-owned natural resources. Under CERCLA, OPA and many state NRD laws, NRD claims are limited to claims for injuries to natural resources owned or under the control of trustees. Claims for natural resource damages on private property are not included. For example, if an oil pipeline ruptures and spills oil into a privately-owned lake, the owner of the lake may have private claims—in tort, for example—but cannot bring an NRD claim against the pipeline owner. Nor could federal or state trustees bring NRD claims on the owner’s behalf. But if the pipeline spill impacted natural resources beyond private property boundaries, for example by injuring migratory birds, trustees could bring NRD claims.

Natural resource damages are non-economic damages. Natural resource damages law is based on a policy judgment that natural resources have non-economic value, albeit value that can be quantified in dollars. They support healthy ecosystems and they provide humans with recreational and aesthetic benefits that have value separate from whatever traditional economic value they may also have. Only compensation for these purportedly non-economic values can be included in an NRD claim.

For example, a hotel owner who operates a private resort on a publicly-owned lake where a pipeline spill occurred would not be able to bring an NRD claim for the reduced bookings or cancellations resulting from the pollution. (Again, however, he might be able to bring a tort-based claim.) But NRD trustees could seek NRD damages arising from those lost bookings, on the theory that those bookings represent a loss of opportunities for people to enjoy the recreational and aesthetic benefits that the lake would have provided. Such claimed benefits could potentially include swimming, fishing or even generally enjoying lakeside views and knowing that the lake was clean. The slope is obviously slippery, and there is not yet enough case law to define it well.

Natural resource damages do not include remediation. One thing that is clear is that NRDs do not include cleanup costs. Under CERCLA and related federal and state laws, parties responsible for pollution may be required to pay for costs to remove pollutants from affected areas, to install engineering controls and/or to install monitoring equipment to ensure that the pollutant abates completely or at least remains below risk-based regulatory thresholds. These activities may improve site conditions and prevent future risks to public health or the environment, but they do not, by themselves, resolve NRD liability.

For example, the pipeline owner in the scenarios above may be ordered under CERCLA to dredge lake sediments to ensure that the pollutant concentrations are below certain risk thresholds. This removal can reduce future risks to humans or wildlife, and it could also reduce future NRD liability by ensuring that natural resource services are not further degraded. But it may also have the effect of further upsetting sediment populations that are important to the lake ecosystem, thus increasing NRD liability even though it meets CERCLA cleanup requirements.

One very important lesson from this: Those preservations of NRD rights typical of CERCLA consent decrees are important. Potential RPs should pay close attention to whether and how their cleanup efforts are credited toward or debited against any NRD liability.

Natural resource damages are not intended to be punitive. NRD claims do not include penalties or fees for violating any environmental or other law. Penalties may be assessed under other authorities for the same incident(s) that give(s) rise to natural resource damage liability, but they are neither part of the NRD calculation nor an automatic credit against NRD liability. Similarly, restoration costs are not to be estimated based on the degree of negligence or culpability of an RP for causing an unpermitted release that results in natural resource injuries.

Natural Resource Damages Assessment and Liability

To determine what NRD injuries have occurred and may continue, and who is required to pay for restoration, trustees often engage in a NRDA. Under CERCLA and OPA, RPs have the opportunity to participate in the NRDAs and contribute resources and expertise toward determining natural resource injury and restoration costs.4 The ultimate determination of NRDA liability, however, remains with the trustees. Indeed, under some NRD statutes trustees may be entitled to a rebuttable presumption in favor of their injury assessment as long as they follow regulatory procedures.5 Such a presumption shifts the burden of persuasion onto the alleged RP defendant if an NRD claim is not settled and the trustees bring an NRD claim in court.

Although some potential RPs will be reluctant to participate in NRDAs because of concern that the participation will further target them or increase the likelihood they will be sued, the potential variation in the magnitude of damage assessments makes this a very risky strategy. NRDAs are rarely mechanically applied or routinized investigations. They frequently involve calculations and judgments relating to natural resource injury made under conditions of significant uncertainty and methodological disagreements. They can also involve significant disagreements about the appropriateness of candidate restoration projects in making the public “whole,” especially where the projects do not have a clear nexus to injured resources. Allowing them to go forward without outside participation may foreclose future opportunities to limit damage claims.

To show that a party is liable for natural resource damages, and for how much, trustees must make a number of findings. First, they must show that a natural resource has been exposed to an unpermitted release of oil or hazardous substance that was caused by the party. Without any exposure to the release, there is no natural resource damage liability. Second, trustees must show that the exposure resulted in an injury, which is defined under CERCLA NRD regulations as a measurable or observable adverse change in the chemical or physical quality of the resource.6 Ordinarily, the adverse change is one that degrades the quality of the ecological or human use services that the resource provides. Third, the trustee must calculate what, if any, restoration projects are necessary to restore injured resources to baseline. For example, these might include creation of wetlands or restoration of fisheries. If restoration of the affected resource is not possible, projects might provide in a new location a level of ecological or human use services equivalent to those lost.

Trustees may use a variety of techniques to determine the spatial and temporal extent of exposure to oil or hazardous substances and the adverse changes and injuries that result from such exposure.7 These may include literature reviews, field studies, laboratory studies or modeling. To determine exposure, for example, trustees may collect data on the amount of hazardous substances released into the environment and then calculate, based on their interpretation of the available scientific literature, what the likely concentration of chemicals are in the affected area and whether and how those concentrations might change over time. Alternatively, trustees may take some samples to measure actual pollutant concentrations in the field and then determine the potential exposure area based on statistical models. Obviously, both calculations and modeling of this sort are subject to challenge, but most effectively as the assessment is proceeding, not later in court.

The variety of available techniques to determine natural resource injury is similarly wide. Trustees may compare toxicology literature values to either calculated or measured exposure concentrations to determine whether an adverse physical or chemical change is likely to have occurred in an exposed population. Alternatively, trustees may expose animals or plants to the pollution under controlled laboratory conditions and then calculate what the likely injuries are in the field based on laboratory responses. Again, there is considerable room for debate as to these methodologies.

To determine service losses, trustees often apply a number of detailed and highly technical equations to translate measures of injury into seemingly quantifiable concepts such as discounted service-acre years (DSAYs), which purport to measure the amount of ecological services lost over a period of time. These equations may include efforts to project adverse changes in survival, reproduction or growth. Such DSAYs—often described as restoration “debits”—are then used as the basis for selecting restoration projects that generate restoration “credits” necessary to replace lost services.

The extent to which trustees rely on empirical data collection and laboratory analyses, rather than modeling and literature-based calculations, will often depend on the complexity of the release, the potentially affected resources, the size and uniformity of the affected area and the level of effort potential RPs make in enforcing rigor. There is no one-size-fits-all answer for every NRDA. If a release is relatively small, the chemical and toxicological properties of the released substance well known and the affected area uniform, trustees may decide that a simple modeling exercise without sampling is more cost-effective than engaging in more detailed data collection, especially if the data collection could delay implementation of restoration projects or if the costs of the investigation could potentially be greater than the costs of restoration. In some cases, there may be little basis for potential RPs to object. But the less trustees rely on empirical data, the greater the possibility that they rely more on conservative, risk-based assumptions that could unrealistically magnify restoration cost liability.

Regardless of how large or small the NRDA is, each step in the NRDA process—exposure analysis, injury assessment and restoration cost calculation—can be very controversial. There is no broad consensus in the scientific community about how best to ensure that injuries are accurately measured and restoration projects adequately compensate for those injuries. Potential RPs should pay close attention to NRDAs long before they are subject to natural resource liability demands, and they should be prepared to challenge technical assumptions and conclusions that may not be fully warranted by the available evidence.

One of the most important and controversial elements of an NRDA is determining the baseline condition against which natural resource injuries and service losses—and NRD liability—are to be measured. Most natural resource damages claims occur in the context of pre-existing and ongoing environmental degradation from a variety of sources. There are few if any truly pristine environments remaining on the planet that human-caused impacts can be clearly measured and distinguished from an undisturbed background. Moreover, baseline conditions can be degraded by any number of human-caused environmental impacts unrelated to the incident of immediate concern, such as co-occurring contamination, historical contamination, global climate change or subsidence. There is often very little empirical data that fully quantifies the extent and degree of pre-release environmental degradation or that accurately characterizes the sources of such degradation. Similarly, natural variability of affected species populations in terms of reproduction, growth and survival, can be broad, erratic or otherwise poorly understood. This, too, makes it difficult to determine whether a particular release, even though well-documented, is a compensable cause of natural resource injury. For example, certain fish populations can vary widely from year to year depending on a number of naturally-occurring environmental conditions. Mere documentation of an adverse change in a natural resource that is coincident to an unpermitted release often will be scientifically insufficient to show that the change was caused by an alleged RP or that, if it was, that the injury had any impact on the overall viability of the population or on the ecological or human use services it provides. The relevant baseline may need to be established from a longer time horizon to determine whether an unpermitted release caused injuries that result in service losses greater than what would be expected from normal historical variation.


Resolution of NRD liability often takes place in the context of difficult methodological disputes and the absence of definitive empirical data linking potential RPs to liability for restoration costs. Potential RPs need to weigh the costs, benefits and risks of supporting or disputing trustee NRDAs at each step of the process. Regardless of the type of pollutant or size or complexity of the affected area, however, companies facing significant exposures need to be engaged in the NRDA process early, and not wait until the eve of litigation. Only in this way can they efficiently manage and address the technical and legal challenges that will inevitably arise.